A Memory Efficient Deep Reinforcement Learning Approach For Snake Game Autonomous Agents
This work addresses memory efficiency for DRL applications on resource-constrained devices, but it is incremental as it adapts existing methods to a specific game.
The paper tackled the problem of high memory and computational requirements in deep reinforcement learning by proposing a modified DRL method with a lightweight CNN and compressed imagery data, achieving similar performance to other methods in the Snake game.
To perform well, Deep Reinforcement Learning (DRL) methods require significant memory resources and computational time. Also, sometimes these systems need additional environment information to achieve a good reward. However, it is more important for many applications and devices to reduce memory usage and computational times than to achieve the maximum reward. This paper presents a modified DRL method that performs reasonably well with compressed imagery data without requiring additional environment information and also uses less memory and time. We have designed a lightweight Convolutional Neural Network (CNN) with a variant of the Q-network that efficiently takes preprocessed image data as input and uses less memory. Furthermore, we use a simple reward mechanism and small experience replay memory so as to provide only the minimum necessary information. Our modified DRL method enables our autonomous agent to play Snake, a classical control game. The results show our model can achieve similar performance as other DRL methods.